Sharif Digital Repository

We propose a method for steganalysis of still, grayscale images using a novel set of features that are extracted from images. This feature set employs the Gabor filter coefficients to train a multi-layer perceptron neural network and a support vector machine classifier. We show that incorporation of the Gabor filter coefficients to the feature sets of images could have a significant role in discrimination between clean and altered images. Experimental results show that the proposed method outperforms previous methods, introduced for steganalysis of LSB-matching image steganography, in terms of both discrimination accuracy and feature set dimensionality. © 2009 IEEE  

A feed forward multi-layer perceptron neural network was developed to predict carbon dioxide loading capacity of chemical absorbents over wide ranges of temperature, pressure, and concentration based on the molecular weight of solution. To verify the suggested artificial neural network (ANN), regression analysis was conducted on the estimated and experimental values of CO2 solubility in various aqueous solutions. Furthermore, a comparison was performed between results of the proposed neural network and experimental data that were not previously used for network training, as well as a set of data for binary solutions. Comparison between the proposed multi-layer perceptron (MLP) network and... 

In this paper we present an adaptive method for image compression that is based on complexity level of the image. The basic compressor/de-compressor structure of this method is a multilayer perceptron artificial neural network. In adaptive approach different Back-Propagation artificial neural networks are used as compressor and de-compressor and this is done by dividing the image into blocks, computing the complexity of each block and then selecting one network for each block according to its complexity value. Three complexity measure methods, called Entropy, Activity and Pattern-based are used to determine the level of complexity in image blocks and their ability in complexity estimation... 

Neural networks have been applied very successfully in the identification and control of dynamic systems. The universal approximation capabilities of the multilayer perceptron have made it a popular choice for modeling nonlinear systems and for implementing general-purpose nonlinear controllers. In this paper we try to model and control the mass-spring-damper mechanism as a 1 DOF system using neural networks. The control architecture used in this paper is Model reference controller (MRC) as one of the popular neural network control architectures  

The authors simulated a reservoir by using two-layer perceptron. Indeed a model was developed to simulate the increase in oil recovery caused by bacteria injection into an oil reservoir. This model was affected by reservoir temperature and amount of water injected into the reservoir for enhancing oil recovery. Comparing experimental and simulation results and also the erratic trend of data show that the neural networks have modeled this system properly. Considering the effects of nonlinear factors and their erratic and unknown impacts on recovered oil, the perceptron neural network can develop a proper model for oil recovery factor in various conditions. The neural networks have not been... 

In this article, I have tried to design an intelligent system which can separate and classify perfect and defective circular plates according to their size. After preprocessing, specifications of defects and size are determined through image processing, and finally, a system is proposed through perceptron neural networks methods, neuro fuzzy method, and Lvq to separate these products on basis of their size and defects. In the designing of this system, when input and its related intend is obvious before training network, perceptron neural networks give more exact results. If input and its related output have been clarified but the output have been related to some sub-inputs, lvq method is... 

In this work, an optimum multilayer perceptron neural network is developed to model the correlation between hot working parameters (temperature, strain rate and strain) and flow stress of IN625 alloy. Three variations of standard back propagation algorithm (Broyden, Fletcher, Goldfarb and Shanno quasi-Newton, Levenberg-Marquardt and Bayesian) are applied to train the model. The results show that, in this case, the best performance, minimum error and shortest converging time are achieved by the Levenberg-Marquardt training algorithm. Comparing the predicted values and the experimental values reveals that a well trained network is capable of accurately calculating the flow stress of the alloy... 

In this paper, a novel approach has been presented to simulate and optimize the pulsating heat pipes (PHPs). The used pulsating heat pipe setup was designed and constructed for this study. Due to the lack of a general mathematical model for exact analysis of the PHPs, a method has been applied for simulation and optimization using the natural algorithms. In this way, the simulator consists of a kind of multilayer perceptron neural network, which is trained by experimental results obtained from our PHP setup. The results show that the complex behavior of PHPs can be successfully described by the non-linear structure of this simulator. The input variables of the neural network are input heat... 

Ultrasonic Testing (UT) is one of the most common types of nondestructive methods that is being used in various industries, especially in the automotive industry. In this paper, qualitative and quantitative control of resistance spot welds on three-sheet joints was studied. Initially, mathematical model of ultrasonic waves was extracted for triple sheet joints. Then, acoustic simulation of ultrasonic testing on spot welds was performed using Finite Element Method (FEM). Afterwards, A Multilayer Perceptron (MLP) neural network was used to classify spot welds based on their diameter. There was a mean error of 20.9 % between peak amplitudes of numerical and theoretical models which the most... 

Oil and gas are likely the most important sources for producing heat and energy in both domestic and industrial applications. Hydrocarbon reservoirs that contain these fuels are required to be characterized to exploit the maximum amount of their fluids. Well testing analysis is a valuable tool for the characterization of hydrocarbon reservoirs. Handling and analysis of long-term and noise-contaminated well testing signals using the traditional methods is a challenging task. Therefore, in this study, a novel paradigm that combines wavelet transform (WT) and recurrent neural networks (RNN) is proposed for analyzing the long-term well testing signals. The WT not only reduces the dimension of... 

The authors studied the efficiency and accuracy of neural network model for prediction of permeability as a key parameter in reservoir characterization. So, some multilayer perceptron (MLP) neural network models with different learning algorithms of Levenberg-Margnardt, back propagation, improved back propagation (IBP), and quick propagation with three layers and different node numbers (3, 4, 5, 6, 7) in the middle layer have been presented. These models have been obtained by 630 permeability data from one of offshore reservoirs located in Saudi Arabia. The accuracy of models was studied by comparing the obtained results of each model with experimental data. So, the neural network with IBP... 

This paper describes a neural network controller for autonomous underwater vehicles (AUVs). The designed online multilayer perceptron neural network (OMLPNN) calculates forces and moments in earth fixed frame to eliminate the tracking errors of AUVs whose dynamics are highly nonlinear and time varying. Another OMLPNN has been designed to generate an inverse model of AUV, which determine the appropriate propeller's speed and control surfaces' angles receiving the forces and moments in the body fixed frame. The designed approximation based neural network controller with the use of the backpropagation learning algorithm has advantages and robustness to control the highly nonlinear dynamics of... 

Interaction quality improvement in a social robotic platform can be achieved through intention detection/prediction of the user. In this research, we tried to study the effect of intention prediction during a human-robot game scenario. We used our humanoid robotic platform, RASA. Rock-Paper-Scissors was chosen as our game scenario. In the first step, a Leap Motion sensor and a Multilayer Perceptron Neural Network is used to detect the hand gesture of the human-player. On the next level, in order to study the intention prediction’s effect on our human-robot gaming platform, we implemented two different playing strategies for RASA. One of the strategies was to play randomly, while the other... 

Multilayer perceptron (MLP) neural networks have been used extensively for estimation/regression of parameters. Moreover, recent studies have shown that learning algorithms of MLP which are based on Gaussian function are more accurate. In this paper, the mass attenuation coefficient (MAC) of gamma radiation for light-weight materials (e.g. O-8), mid-weight materials (e.g. Al-13), and heavy-weight materials (e.g. Pb-82) is modelled using Gaussian function based regularization of MLP (i.e. Bayesian regularization (BR)) and by a modular estimator. The results are compared with the Reference results. To show better performance of the utilized algorithm, the results of the different supervised... 

One of the most important challenges in target parameters estimation via model-free methods is selection of the most effective input parameters namely features selection (FS). Indeed, irrelevant features can degrade the estimation performance. In the current study, the challenge of choosing among the several plant parameters is tackled by means of the innovative FS algorithm named ranking of features with minimum deviation from the target parameter (RFMD). The selected features accompanied with the stable and the fast learning algorithm of multilayer perceptron (MLP) neural network (i.e. Levenberg-Marquardt algorithm) which is a combination of gradient descent and Gauss-newton learning... 

In this paper, an efficient stress recovery technique is presented to estimate the recovered stress field at the nodal points. The feed–forward back–propagation multilayer perceptron (MLP) neural network approach is employed to improve the accuracy of the stress recovery method. An automatic adaptive mesh refinement is performed based on a–posteriori Zienkiewicz–Zhu error estimation method. The proposed technique is employed to recover the stress field accurately in the regions with a high stress gradient where the conventional recovery techniques are not able to improve the stress fields efficiently due to the singular behavior of problem. Finally, several numerical examples are solved to... 

Viscosity of mixtures of biodiesels (admixtures) and mixtures of biodiesel/diesel (blends) is a important parameter for determining their combustion behavior. There is no universal and general model for prediction of viscosity of these systems at different conditions. Hence, developing simple, accurate, and general models for prediction of viscosity of these systems is of great importance. In this work, three computer-based models named multilayer perceptron neural network (MLP-NN), radial basis function optimized by particle swarm optimization (PSO-RBF), and adaptive neuro fuzzy inference system optimized by hybrid approach (Hybrid-ANFIS) were developed for prediction of viscosity of blends... 

This paper illustrates the use of a combined neural network model for classification of electrocardiogram (ECG) beats. We present a trainable neural network ensemble approach to develop customized electrocardiogram beat classifier in an effort to further improve the performance of ECG processing and to offer individualized health care. We process a three stage technique for detection of premature ventricular contraction (PVC) from normal beats and other heart diseases. This method includes a denoising, a feature extraction and a classification. At first we investigate the application of stationary wavelet transform (SWT) for noise reduction of the electrocardiogram (ECG) signals. Then...